Epidemiological risk factors of suicidal behavior and effects of the components of coping strategies on suicidal behavior in medical students: An institution-based cross-sectional study in India.

Background: Suicidal behavior in medical students can be triggered by mental stresses and adoption of poor coping strategies, and might have a negative impact on their quality of life. Aims: The aim of this study was to evaluate the prevalence and risk factors of suicidal behavior, and effects of the components of coping strategies on suicidal behavior among medical students.
Methods: An institution-based cross-sectional study was conducted on 531 medical students for a period of two months from February to March 2021. Stratified random sampling technique was used to select the study participants. Data was collected using a self-administered questionnaire. Suicide Behaviors Questionnaire-Revised (SBQ-R) scale was used to measure suicidal behavior. Exploratory factor analysis was performed on the Brief-COPE Inventory to classify the coping components. Chi-squared test and multiple logistic regression were used to determine the risk factors and their association with suicidal behavior.
Results: A total of 104 respondents (19.6%) had reported an SBQ-R cutoff score of ≥7 and had suicidal behavior. The rate of lifetime suicidal ideation, plan and attempt was 20.3%, 10.3% and 2.3%, respectively, among medical students with one-year prevalence of suicidal ideation at 33%. The mean age of the participants was 21.26 years (standard deviation (SD) = 1.99). The identified risk factors significantly associated with higher suicidal behavior were depression (OR = 9.6), dissatisfaction with academic performances (OR = 4.9) and coping with mental disengagement (OR = 4.6), while coping with supportive strategies (OR = 0.57) was investigated as a preventive factor for the suicidal behavior. Conclusions: The prevalence of suicidal behavior is found to be highly alarming, revealing that depression and poor coping strategies are risk factors among medical students. The analysis recommends that quantification of the problems, treatment at an early stage and proactive student counselling to help them embrace the appropriate coping strategies should be the first steps in prevention of suicidal behavior. Copyright:

INTRODUCTION

Suicidal behavior is a term incorporating suicidal ideation and its consequences, including suicidal thoughts, plans, attempts and completed suicide.[1] It is a remarkable contributor to the global burden of disease, resulting from complex interconnections of bio-psycho-sociological along with environmental and cultural factors.[2] Suicide rates are highest amongst young people (15–39 years), and ranks first and second amongst the causes of mortality in women and men, respectively.[3] The literature concluded that medical students or workforce of the future are at higher risk of suicidal behavior as compared to students of other courses and the general population.[4] Suicidal behavior, despite being a paramount public health crisis, is one of the unrecognized concerns in the society of younger medical students.[5]

It was lamentable for the medical students that they started facing numerous mental stresses at the beginning of their medical studies and became overburdened until their transformation into young skilled physicians, which exacerbated the greater psychological problems and eventually made them highly vulnerable to depression.[67] Medical students with depression or poor mental health with poor coping strategies might desert their profession, which might lead to suicidal behavior.[8910] Recent studies revealed that 1.3%–32.7% and 1.8%–53.6%, of medical students experienced suicidal behavior[11] and life-time suicidal ideation, respectively.[12] It was found that medical students are under overwhelming mental stress, not only due to academic or professional issues, but also due to chronic non-academic issues like maladaptive personality and stressful life incidents.[13] Previously, it was also established that the danger of suicidal behavior increased manifold in medical students with family history of mental illnesses, having had suicidal ideation,[14] pre-existing psychiatric disorders,[15] dissatisfaction with life, decreased self-esteem and decreased engagement in social dealings with family, friends or colleagues.[1617] Due to these stresses, one would expect a higher level of suicidal ideation in Indian medical students. However, there is limited data available regarding suicidal behavior and its risk factors in such students.[51718] A recent study conducted in India[18] found that 29.6% of medical students had suicidal behavior while a previous study showed a higher proportion (53%).[5] Given the limited amount of research conducted, more information on suicidal behavior and its associated risk factors in Indian medical students is necessary as their recognition at an appropriate time can help mitigate it and stage prompt interventions.[19]

Coping strategies are dynamic processes involving cognitive and behavioral efforts that individuals use to conceptualize their thoughts, feelings and actions encountered during various stressful conditions, with considerable inter- or intra-individual variability.[2021] There has been concordance that, appropriate coping strategies estimates favorable future outcomes, encompasses higher degree of ego development and infrequent behavioral problems, higher self-esteem, fewer psychological problems and negates suicidal behavior.[2223] Few studies consistently demonstrated that appropriate coping styles could generate positive emotions and act as a deterrent to suicidal behavior even in the presence of stress,[2425] while ineffective coping skills could produce unfavorable emotions and increase the risk of suicidal behavior during confronted stressful situations.[26] Furthermore, studies on students suggested that maladaptive coping skills with the objective of maintaining well-being (such as reducing the negative emotions related to stressors and temporarily alter the focus from stressful conditions to other important things) could arrest the suicidal behavior.[2527] However, the diverse coping skills of medical students to alleviate suicidal behavior is emphatically pronounced across the globe,[42728] but the available research in conceptualizing and classifying the distinct coping strategies is inconsistent.[2829] To the best of our knowledge, none of the previous studies have evaluated the suicidal behavior and its risk factors, and different components of coping strategies and their effects on suicidal behavior in medical students in the same study sample, especially in the Indian context. Therefore, the present study has endeavored to assess the magnitude of suicidal behavior and the risk factors affecting it, and classify the dimensions of coping strategies and their influence on suicidal behavior in medical students in an institution located in Northern India.

METHODOLOGY

Study design and settings

This was a descriptive cross-sectional study, performed on undergraduate and postgraduate medical students of an institution located in North India, conducted from 2 February 2021 to 29 March 2021. Every year, around 100 undergraduates (except first and second academic years where 120 students were enrolled) and 40 postgraduate medical students get admitted to the college. Currently, a total number of 660 medical students are studying in this institution. This study was carried out after getting ethical approval from the Institutional Ethical Committee Board and in accordance with Ethical Committee standards and the Helsinki declaration. During the study, the anonymity and confidentiality of the participants was assured, and was maintained as their personal information, like name or contact was not asked.

Sample size

The study’s required sample size (N = 531) was calculated by using single population proportion formula. It was calculated on the basis of the following assumptions: the prevalence or expected proportion of suicidal behavior (P) is 29.6% as evaluated in a previous survey;[18] the absolute precision is 4% (d) at 95% confidence interval (Z) where the value of Z is 1.96 (constant) and non-response rate was also considered to be 5%.

Study sample

Students from all batches of undergraduate and postgraduate courses aged 18 years or older (both male and female), able to read and understand English and willing to give informed consent were included in the study while medical officers, house-officers and consultants were not included in the study. The study excluded participants who did not provide informed consent to participate in the survey.

Sampling and data collection procedure

A stratified random sampling method was applied to make the strata of the students of each academic year, and then the total sample size was allocated proportionately to each academic year of undergraduates (1 to 4 and internship) and postgraduates. Finally, computer-generated random number table was used to select and enlist each study participant from different academic years (97 each from 1st and 2nd year, 80 each from 3rd, final year and internship batch, and 97 from postgraduate batch). The purpose of the study and importance of honest answers were briefed to the participants, and privacy and confidentiality of their information was also assured. Then, questionnaire sheets were distributed to the selected participants by hand in their classrooms before lectures and during breaks, and written informed consent was obtained from them before eliciting the required information. At the end of the description, helpline numbers and email address were provided for those in need of professional help.

Data collection measures

The data was collected by using the self-administered questionnaires which had six sections (A to F), consisting of brief information regarding the study purposes in section A, written informed consent in section B and about the basic information of students in section C. Section D had Patient Health Questionnaire-9 (PHQ-9) to measure depression in students. Four questions related to Suicide Behaviors Questionnaire-Revised (SBQ-R) scale which measured suicidal behavior, formed section E of the questionnaire. The last part of the questionnaire (section F) had the coping inventory to analyze the coping strategies adopted by students.

Basic information: This section of the survey had seven questions regarding sociodemographic characteristics. Sociodemographic data was collected in the form of current age, gender (male/female), residence (urban/rural), year of study, satisfaction with their academic performance (satisfied/unsatisfied/can’t say), and about their current tobacco smoking and current alcohol consumption status (yes/no). Current use in this study was defined as consuming specific substances within the last 3 months.[14]

Patient Health Questionnaire-9 (PHQ-9) scale: PHQ-9 proved to be a very useful tool to detect the levels of depression, it being the main reason of selecting this tool for the present study.[30] PHQ-9, a 9-item depression module, is easy to administer. Participants were asked over the last two weeks how often they had been bothered by the depressive symptoms. Each item is scored from 0 to 3 (0 = not at all, 1 = several days, 2 = more than half the days, 3 = nearly every day). A total score ranges from 0 to 27. The level of depression is categorized as minimal (0–4), mild (5-9), moderate (10-14), moderately severe and severe (>20). In the context of the present study, during the analysis, two groups were created by using score <10 for non-depressed group and ≥10 for depressed group, because the cut-off score of 10 had shown good sensitivity (82%) and specificity (93%) in a structured diagnostic interview.[31] The Cronbach’s alpha of scale in the present study was 0.884, presenting good internal consistency reliability.

Suicide Behaviors Questionnaire-Revised (SBQ-R): SBQ-R is a very useful tool to measure the different dimensions of suicidality, validated firstly by Osman et al.[32] This scale helps in covering the broad range of information in a very brief administration. This scale has good reliability and validity for psychiatric and non-psychiatric populations. It has four items that evaluate lifetime suicidal ideation, suicidal plan and suicide attempt (Item-1 with 4 points, cutoff score ≥2), the frequency of suicidal ideation over the past 12 months (Item-2 with 5 points, cutoff score ≥2), the threat of suicide attempt (Item-3 with 3 points, cutoff score ≥2) and self-reported likelihood of suicidal behavior in the future (Item-4 with 6 points, cutoff score ≥3). In the present study, item-1 question # 2 for lifetime suicidal ideation; item-1 question # 3a and 3b for suicidal plan; and item-1 question # 4a and 4b for suicidal attempt were used. A total calculated score of ≥7 was considered as the cutoff score for the students to have suicidal behavior, had shown very good sensitivity (93%) and good specificity (95%) in a structured diagnostic interview. Responses can be used to identify at-risk individuals and specific risk behaviors. The Cronbach’s alpha of scale in the present study was 0.801, presenting good internal consistency reliability.

The Coping inventory: The coping mechanisms adopted by the students were assessed by using the selected items of the Brief-COPE Inventory. It consists of 14 coping mechanisms. Each dimension contains two items. The instruction related to coping mechanisms in the scale was like “these items deal with ways you have been coping with the problems in your life…”. These items are rated using a 4-point Likert-type scale from 1 to 4 with 1 being “I have not been doing this at all” and 4 being “I have been doing this a lot.” In order to minimize the burden, two of the authors combined both items and collected one item for each coping strategy using a consensus procedure. Then, a pre-test was conducted using this modified inventory on 25 students (5% of total sample size). None of these students faced any difficulty in either understanding or answering the questions. Cronbach’s alpha for this modified inventory was 0.842, with a range from 0.823 to 848, showing good internal consistency and reliability while the original Cronbach’s alpha for the Brief-COPE scale ranged from 0.623 to 0.926.[33] Finally, these 14 items representing each coping strategy were chosen for the self-administered questionnaire of the present research.

Statistical analysis

The data was entered and analyzed using Statistical Package for the Social Sciences (SPSS) version 25.0 (IBM, Chicago, IL, USA). Cronbach’s alpha was used to assess the internal consistency of the scales. Exploratory factor analysis (EFA), using varimax rotation and principal component analysis, was performed to unmask the components of these 14 coping items and to assess the construct validity. Factor loadings value of ≥ 0.5 was set for the variables to be a contributing variable in the factor analysis. Categorical variables were calculated as frequencies and percentages, and were compared via the Chi-squared test (or Fisher’s exact test where frequency in any cell was less than 5). Continuous variables were calculated as mean and standard deviations, and were compared using independent student t test. Pearson’s correlation was used to find out the correlation between suicidal behavior and designated dimensions of coping strategies. Finally, multivariable logistic regression analysis was applied to find out the independent factors by adjusting the variables significant in univariate analysis, and adjusted odds ratio (AORs) and confidence interval (95% CI) were used to evaluate the strength of association between independent factors and suicidal behavior. Statistically significant level was set at P < 0.05 (two-tailed).

RESULTS

Description of Socio-demographic characteristics of study participants

Five hundred thirty-one medical students were enrolled in the present study. Out of them, a majority of medical students (85%, 449) were female and the remaining were male. The mean age of the participants was 21.26 years (standard deviation = 1.99), with the age ranging between 18 and 30 years. Most of the participants (60%) were in the age group of 21–25 years followed by age group of 18–20 years (35%). Around more than half of them belonged to rural areas as showed in Table 1. Table 1 also illustrates that more than two-third (69%) of the medical students were dissatisfied with their academic performance and the rest were satisfied. Additionally, almost all of the participants, that is, 95.5% and 90% reported that they were not currently using tobacco and alcohol, respectively. Around 55.2% of medical students were free from any depressive symptoms. A large number of medical students (44.8%) were moderate-to-severely depressed (PHQ-9 score ≥10).

Table 1

Sociodemographic characteristics of medical students with prevalence of suicidal behavior in subgroups

Variables	Subgroups	Numbers (%)
		Total Medical Students n=531 n (%)	Non-Suicidal Group (SBQ-R score <7) n=427 n (%)	Suicidal Group (SBQ-R score ≥7) n=104 n (%)	P
Age (in years)	18-20	187 (35.2)	146 (34.2)	42 (40.4)
	21-25	318 (59.9)	260 (60.9)	58 (55.8)	0.601
	More than 25	26 (4.9)	21 (4.9)	4 (3.8)
Gender	Male	82 (15.5)	67 (15.7)	15 (14.4)	1.000
	Female	449 (84.5)	360 (84.3)	89 (85.6)
Residence	Urban	242 (45.6)	195 (45.7)	47 (45.2)	1.000
	Rural	289 (54.4)	232 (54.3)	57 (54.8)
Satisfaction with academic performance	Dissatisfied	365 (68.7)	265 (62.1)	100 (96.1)
	Satisfied	156 (29.3)	152 (35.6)	4 (3.9)	<0.001**
	Can’t say	10 (1.9)	10 (2.3)	0
Current tobacco smoking status	Yes	24 (4.5)	17 (3.9)	7 (6.7)	0.288
	No	507 (95.5)	410 (96.1)	97 (93.3)
Current alcohol consumption status	Yes	51 (9.6)	39 (9.1)	12 (11.5)	0.459
	No	480 (90.4)	388 (90.9)	92 (88.5)
Depression	Non-depressed group (PHQ-9 score <10)	293 (55.2)	275 (64.4)	18 (17.3)	<0.001**
	Depressed group (PHQ-9 score ≥10)	238 (44.8)	152 (35.6)	86 (82.7)

**P significant is <0.05 and highly significant is P<0.001

Prevalence of suicidal behavior (suicidal ideation/attempt/plan) on SBQ-R and its association with socio-demographic variables

A small number of respondents (104, 19.6%) had reported suicidal behavior (SBQ-R score ≥7) while the remaining 427 medical students having SBQ-R score <7 were included in the non-suicidal group. The rate of lifetime suicidal ideation, plan and attempt was 20.3%, 10.3% and 2.3%, respectively, among medical students. A higher number of participants (179, 33.7%) had suicidal ideation over the past one year. Around 101 students had threatened to attempt suicide more than once. A minimum number of study participants (48) revealed self-reported likelihood of suicidal behavior in the future. Mean score on SBQ-R was 4.97 (standard deviation = 2.80). The prevalence of suicidal behavior did not vary significantly among the sociodemographic factors, although it was found to be highest in female and in students aged 18–20 years. Chi-squared analysis showed that a larger percentage of students in the suicidal group were significantly dissatisfied with their academic performance (96% vs 63%). The rate of suicidal behavior also varied significantly in students who had moderate-to-severe depressive symptoms (83% vs 35%). The study did not reveal any significant effect of current use of substance consumption behavior (tobacco/alcohol) in students on their suicidal behavior [Table 1].

Factor loadings of coping strategies and their psychometric properties

EFA on the 14 items of coping strategies yielded four potential latent factors explaining 70.5% of total variance, with satisfactory eigenvalue (>1) and factor loadings (≥ 0.5). The Kaiser–Meyer–Olkin (KMO) measure of sampling adequacy (0.856) and the Bartlett’s test of sphericity (Chi-squared test = 2702.274; df = 91; P < 0.001) showed that the sample size for factor analysis was adequate.[34] The four factors or components identified from the EFA and principal component analysis (PCA) followed by varimax rotation with Kaiser normalization assess the participant’s level of coping strategies with respect to (a) psychological resilience (factor 1) comprising of five coping items (Q1, Q4, Q5, Q6, Q9) which explained 34.7% of the variance; (b) supportive strategies (factor 2) comprising of two coping items (Q2, Q3) which explained 15.3% of the variance; (c) religiosity and humor (factor 3) consisting of two coping items (Q7, Q8) which explained 12.2% of the variance; and (d) mental disengagement (factor 4) consisting of five coping items (Q10, Q11, Q12, Q13, Q14) which explained 8.3% of the variance. The internal consistency (Cronbach’s alpha) of each factor was 0.84, 0.79, 0.83, and 0.70, respectively, evaluating the excellent reliability as the value of Cronbach’s alpha in this study was found to be more than 0.7. The reliability analysis also illustrated that all of the 14 items were to be retained, as shown in Table 2. For further analysis, each item of the four factors were summed up to derive continuous variables (with a range from 5 to 20 for factor 1 and factor 4, and range from 2 to 8 for factors 2 and 3) and two binary variables (with cutoffs at ≥15 and ≥,6 respectively), corresponding to a mean value of at least a medium amount of time of using the corresponding coping strategies by medical students.

Table 2

Rotated structure matrix for principal component analysis with varimax rotation of 14-item Brief-COPE Inventory, showing loadings value of different factors (major loadings ≥0.5)

Sr. No.	Statements	Factor 1	Factor 2	Factor 3	Factor 4	Cronbach’s Alpha Coefficient if Items Deleted
Q1.	I have been concentrating on my efforts/taking action to make the situation better.	0.725				0.829
Q2.	I have been getting emotional support/comfort from others.		0.799			0.824
Q3.	I have been getting help and advice from other people.		0.805			0.826
Q4.	I have been trying to see/looking good in what is happening.	0.696				0.823
Q5.	I have been thinking to come up with a strategy about what to do.	0.729				0.824
Q6.	I have been accepting the reality and learning to live with fact that it has happened.	0.783				0.826
Q7.	I have been trying to find comfort in religious or spiritual beliefs like praying or meditation.				0.596	0.825
Q8.	I have been making fun of the situation.				0.574	0.824
Q9	I have been turning to other activities such as movies, sleeping, watching TV, reading, shopping and other works to take my mind off the things.	0.739				0.825
Q10.	I have been refusing to believe that it is real.				0.567	0.836
Q11.	I have been using alcohol or other drugs to get rid of it and feel better.				0.571	0.845
Q12.	I have been giving up the attempts to deal/cope with it.				0.750	0.846
Q13.	I have been expressing my negative feelings and saying to let my unpleasant feelings escape.				0.687	0.835
Q14.	I have been blaming/criticizing myself for the things that have happened.				0.754	0.848
	Eigenvalue	4.865	2.136	1.716	1.160
	% of variance	34.747	50.007	62.266	70.555
	Cronbach’s alpha coefficient	0.843	0.799	0.830	0.704

Extraction Method: Principal component analysis with varimax rotation with Kaiser normalization

Components of coping strategies adopted by medical students and their correlation with suicidal behavior

Mean score of each component of coping strategy along with significant differences in suicidal and non-suicidal groups are shown in Table 3. Supportive strategies were significantly used by students in the non-suicidal group (5.33 vs 4.70) while mental disengagement was significantly used by students in the suicidal group (11.89 vs 8.99). Correlational analysis between these two variables is also depicted in Table 3. Suicidal behavior among respondents was weakly negatively correlated to supportive strategies (r = −0.165, P < 0.001) which explained that more use of this component was associated with lower prevalence of suicidal behavior. Additionally, suicidal behavior was moderately positively correlated to mental disengagement (r = 0.413, P < 0.001) which explained that more use of this component was associated with higher prevalence of suicidal behavior in medical students [Table 3]. Psychological resilience (80%), and religiosity and humor (61%) were the more often used coping strategies in the study population as compared to supportive strategies (47%). Mental disengagement (denial, substance use, venting, behavioral disengagement and self-blame) was a rarely used coping strategy (18%) in medical students, as shown in Figure 1.

Table 3

Comparison of the responses to the components of coping strategies used by medical students and their correlation with suicidal group

Coping Strategies	Total Mean (SD)	Non-Suicidal Group Mean (SD)	Suicidal Group Mean (SD)	P	Suicidal Group (r)
Factor 1 (Psychological resilience)	15.21 (3.65)	15.28 (3.82)	14.93 (2.84)	0.381	−0.040
Factor 2 (Supportive strategy)	5.20 (1.88)	5.33 (1.86)	4.70 (1.90)	< 0.05*	−0.165 **
Factor 3 (Religiosity and humor)	5.80 (1.58)	5.85 (1.61)	5.58 (1.48)	0.109	−0.096
Factor 4 (Mental disengagement)	9.56 (3.36)	8.99 (3.07)	11.89 (3.54)	< 0.001**	0.413**

*P significant; **P highly significant; = Pearson correlation coefficient

Figure 1

Components of coping strategies adopted by medical students

Effects of variables and components of coping strategies on suicidal behavior

The statistically significant variables (dissatisfaction with academic performance, depression, supportive strategy and mental disengagement) in Chi-squared analysis were assigned for adjusted multivariable logistic regression analysis and several factors were determined which were associated with potential risk of suicidal behavior along with protective factors. The value of R-square (0.386) showed that the whole model explained 38.6% of variance in suicidal behavior. The most evident result was that the students who were dissatisfied with academic performance had significantly greater risk of suicidal behavior (AOR = 9.618, 95% CI = 3.351–27.600; P < 0.001) than those who were satisfied with their academic performance. The only protective factor for suicidal behavior was the use of supportive strategies (AOR = 0.577, 95% CI = 0.321–1.057; P < 0.05). The common significant variables in students were having depression (AOR = 4.960, 95% CI = 2.760–8.915; P < 0.001) and coping through mental disengagement (AOR = 4.630, 95% CI = 2.561–8.371; P < 0.001), though the odds of developing suicidal behavior in any of these factors were around 5, as showed in Table 4.

Table 4

Multivariable logistic regression on correlates of suicidal behavior among medical students

Variables	Suicidal Behavior AOR (95% CI)
Dis-satisfaction with academic performance.
No	1
Yes	9.618 (3.351-27.600)**
Depression
No	1
Yes	4.960 (2.760-8.915)**
Psychological resilience
No	1
Yes	1.250 (0.600-2.602) 0.551
Supportive strategies
No	1
Yes	0.577 (0.421-1.057)*
Religiosity and humor
No	1
Yes	0.651 (0.358-1.183) 0.159
Mental disengagement
No	1
Yes	4.630 (2.561-8.371)**

AORs represent adjusted odds ratios. CI represents confidence interval. *P significant, **P Highly significant. Model parameters for suicidal behavior: Cox and Snell R2=0.243, Negelkerke R2=0.386

DISCUSSION

This descriptive cross-sectional study in an institution located in North India observed that one-third of medical students experienced suicidal ideation, and one-fifth had suicidal behavior along with few risk factors (dissatisfaction with academic performance and depression) contributing to it, and also observed the effects of components of coping strategies on risk of suicidal behavior. However, exclusive evaluation of suicidal behavior might help in bolstering the well-being of medical students.

Prevalence of suicidal behavior (suicidal ideation/attempt/plan) and its association with sociodemographic variables

In the present study, the magnitude of suicidal behavior (SBQ-R ≥7) among medical students was found to be 19.6%, suggesting a higher rate among them, and may require wellness curricula in medical colleges to address and mitigate these distresses. Regarding the magnitude of suicidal behavior, these findings are in assonance with the observations found in studies done among medical students using Composite International Diagnostic Interview (CIDI) in Ethiopia[14] and using SBQ-R in China[28] where the prevalence rate was reported at 20.6% and 21.4%, respectively. The reasons for such high prevalence of suicidal behavior could be due to the problems related to studies, families and friends, and self-centered behavior in the medical students. In contrast to the present study, much higher prevalence of suicidal behavior, of 29.6% and 28.9%, on SBQ-R was estimated in studies conducted in India[18] and Ethiopia,[35] respectively, while another study conducted in Malaysia[36] revealed a much lower prevalence (7%) rate as compared to the present study. These discrepancies might be related to differences in sociocultural background, sample size and study design used.

In the present study, the magnitude of suicidal ideation over the past one year and lifetime suicidal ideation among medical students was found to be 33.7% and 20.3%, respectively. These findings were in line with a few cross-sectional surveys done in Pakistan,[15] Ethiopia[35] and Belgium[37] in context to suicidal ideation over the past one year, and also, with ones conducted in India[18] and China[38] in context to lifetime suicidal ideation. In contrast to the present study, much higher prevalence of suicidal ideation over the past one year was estimated (53%) in a study conducted in India by Goyal et al.[5] However, a few studies conducted in Ethiopia[39] and Oman[40] revealed suicidal ideation over the past one year of 23% and 21%, respectively, which were much lower than the present study. Additionally, the findings regarding lifetime suicidal ideation observed by a few cross-sectional studies done in Ethiopia (58%)[35] and Oman (33%)[40] were inconsistent with the prevalence rate revealed in the present study. In other words, the prevalence of suicide plan (10.3%) and attempts (2.3%) in the present study were lower compared to recent studies in India (12.5% and 5.4%)[18] and Ethiopia (37.3% and 4.4%).[35] Such heterogeneity in the observation could be attributed to diversities in used measurements for suicidal ideation like Patient Health Questionnaire and Beck Suicidal Questionnaire which assesses the suicidal ideation for the past two weeks only, and differences in economic growth, legal system, attitudes and values.[41]

Suicidal behavior in the present study population was not influenced by sociodemographic factors (age, gender, place of residence and tobacco or alcohol consumption behavior), similar to a study done by Adhikari et al.[42] and Torres et al.[43] This result demonstrates that medical students in every age group, either male or female, might have faced similar negative emotions and problems during their medical education. Contrary to the present study, a few cross-sectional surveys[514] pointed out that suicidal ideation was significantly higher in female medical students than their male counterparts. Recently, a few authors from Ethiopia[39] and Portugal[44] established a significant association where alcohol consumption behavior increased more than two times the odds of suicidal behavior among medical educators, which is incongruent to the findings observed in the present study. Among the sociodemographic variables, only dissatisfaction with academic performance was significantly associated with suicidal behavior in participants, which is a finding precisely similar to observations revealed by studies done in Asian countries like Pakistan,[15] China[28] and India.[4546] The present study investigated that the propensity of suicidal behavior increased ten times in medical students who were dissatisfied with their academic performances than their counterparts, which was identified as the strongest risk factor of such behavior. This result is in concordance with research showing dissatisfaction with academic performance to be a significantly strong risk factor (2–3 times the odds) for suicidal behavior.[1528] The present study evaluated that the second strongest predictor for suicidal behavior in medical students was depression (found in nearly half of the medical students). Respondents with depressive symptomatology (PHQ-9 score ≥ 10) had 4.9 times higher odds of suicidal behavior than those without depression. An analytical cross-sectional study conducted in Malaysia on 657 medical students also established depression as one of the strongest predictors of suicidal behavior (5.9 times higher the odds).[36] Another institution-based cross-sectional study showed that depressed students were 10.1 times more likely to have suicidal behavior than their counterparts.[14] Abdu et al.[35] observed that multivariable logistic analysis did not confirm the evidence regarding the significant association between suicidal behavior and depression in students. Dissatisfaction with academic performance revolves around extensive work load due to conscientiousness of medical curriculum and poor performance in academics, leading to fear of failure in exam and doubts about academic competence, due to which students might have feelings of worthlessness, hopelessness and uselessness that ultimately lead to depression, which could be the possible explanation for increasing risk of suicidal behavior in these students.[3947]

Coping strategies and suicidal behavior

This was the first attempt in assessing the possible association between suicidal behavior and the components of coping strategies in medical students. In the present study, it was observed that psychological resilience was the most common coping strategy used by medical students, followed by religiosity and humor, and supportive strategies was found to be a significant protective factor while mental disengagement was found to be a significant risk factor for suicidal behavior.

In the present study, psychological resilience was used by a higher proportion of medical students in the non-suicidal group. A qualitative study[48] reported that a spectrum of psychological factors (passive acceptance, active coping and positive reframing) were the potential contributing factors of psychological resilience, as evaluated in the present study. Previous research among medical students have shown that enhancing psychological resilience could buffer the influence of negative emotions and have positive effects on psychological well-being as it is a dynamic process involving outcomes, attributes or process of coping with, adapting to and rebounding from adverse stressful events.[4950] Recently, few studies[5152] identified that psychological resilience was a significant protective factor against the risk of suicidal behavior in high-risk individuals, which is inconsistent with the findings of the present study in which no significant association was revealed between suicidal behavior and psychological resilience. Medical students accepting the problems as a challenge and solving them by distracting themselves with other activities by setting a positive path cognitively could explain the insignificant association between suicidal behavior and psychological resilience. Although having religiosity and humor affiliation was not statistically significant in non-suicidal and suicidal behavior groups, in contrast to expectations, it was found that religiosity and humor as a coping strategy was used by non-suicidal group of medical students in higher proportion. It was noted that in the Brief-COPE scale, the item under “religion” also includes spirituality, and under “humor”, it was not clearly indicated about the use of either affiliated or self-defeating humor, which could be the possible justification for these items to not be presented as a protective factor against suicidal behavior. Eskin et al.[53] also found that those who were spiritual but not religious might not have experienced the protection against suicidal behavior. This non-significant association with suicidal behavior, similar to the present study, was only observed in a few cross-sectional studies where religiosity and humor were evaluated as a separate entity.[2853]

Interestingly, the present study established a significant association between lower risk of suicidal behavior and supportive strategies. In multivariate analysis, students who used supportive strategies had more than half time (AOR = 0.57) lower odds of suicidal behavior than those who did not use this component of coping strategy while facing stressful situations. This finding might have suggested that medical students adopted supportive strategies in response to emotional distress and during the lack of social supportive system and used supportive strategies as an escape mechanism from negative emotions to alleviate the suicidal risk. This could be the possible justification for supportive strategies in being a protective factor against suicidal behavior. Previously in the literature, it was reported that social and emotional support were the protective factors that attenuated the probability of evolution of suicidal behavior in risky cases.[5455] It was revealed that mental disengagement was the only component among the coping strategies more likely to be used by the suicidal group and significantly associated with the higher risk of suicidal behavior among medical students. In multivariate analysis, students who practiced mental disengagement strategies had nearly five times higher odds of developing suicidal behavior than those who did not exert this strategy. These results reflected that students with a high risk of suicidal behavior could have poor problem-solving skills and negative enthusiasm against stressful situations, and symbolized the mental disengagement as a maladaptive strategy to cope with the situations. In assonance with these notions of the present study, a study in China by Tang et al.[25] also established that students with higher risk of suicidal behavior were found to have increased reliance on maladaptive strategies when they had insufficient resolution during acute stressful incidents and sustained exposure to stressful state of affairs. Consistent with the present study findings, maladaptive strategy as a coping mechanism is ineffectual and worsens the mental well-being of students, and is also a predictor of suicidal behavior.[428]

The main strength of the present study is that this study is, to our knowledge, the first to assess the predictive effects of and facilitates the understanding of different components of coping strategies on risk of suicidal behavior in medical students by using EFA method. Secondly, the study also helped in finding vulnerable groups of medical students by using standardized validated tools with very good internal reliability. Thus, the results observed were intriguing and had effective therapeutic implications in the prevention of suicidal behavior among medical students. Lastly, lifetime suicide ideation, suicide plan and suicide attempt were also analyzed separately in the present study.

The present study also has several limitations. Firstly, this study was cross-sectional and limited to a single institute, so, it is not possible to generalize its findings. Such a study design did not allow the investigation of causal association of outcomes and long-term impact of stressful events on suicidal behavior. Secondly, this study did not include psychological issues other than depression, like anxiety or stress, which has been associated with suicidal behavior. This study did not include the other variables like stressful events related to their college life, dependence of substances other than tobacco and alcohol, treatment obtained and past or family history of psychiatric illness or suicidality, which might have influenced the suicidal behavior. Thirdly, information, selection, recall and response bias due to the use of self-administered questionnaire cannot be ruled out in the present study, which is also one of its major limitations. Finally, prevalence and risk factors for suicidal behavior in each academic year were not analyzed separately.

CONCLUSION AND FUTURE SUGGESTIONS

This study concluded that a high proportion of medical students presented with suicidal behavior. It provides empirical evidence that depression and dissatisfaction with academic performance in students along with the use of mental disengagement as a coping strategy were risk factors while supportive strategy was the preventive factor for suicidal behavior. It conveys that medical colleges should make concerted efforts early on to perceive depression along with suicidal behavior among medical students and plan accordingly to strengthen their mental health by organizing counseling sessions, seminars and workshops at the institutional level. Medical educators should provide systematic support to encourage students which might help in facing stressful situations during medical education. The government should aim at integration of stress coping strategies and wellness medical curricula to counter academic pressure, which might help in production of healthy physicians.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.